Electoral College Prediction Model Predicts a Romney Win

"Two University of Colorado professors, one from Boulder and one from Denver, have put together an Electoral College forecast model to predict who will win the 2012 presidential election and the result is bad news for Barack Obama. The model points to a Mitt Romney victory in 2012. Ken Bickers from CU-Boulder and Michael Berry from CU-Denver, the two political science professors who devised the prediction model, say that it has correctly forecast every winner of the electoral race since 1980. "Based on our forecasting model, it becomes clear that the president is in electoral trouble," Bickers said in a press statement." So should we care about this? Or are the circumstances of this election cycle make such models less useful than before?

http://www.huffingtonpost.com/2012/08/22/university-of-colorado-pr_n_1822933.html?icid=maing-grid7|aim|dl1|sec1_lnk2&pLid=195989



• [quote]The model also suggests that Romney will win every state currently considered a swing state which includes Florida, Virginia, North Carolina, Ohio, Pennsylvania, Wisconsin, Minnesota, New Hampshire and Colorado. What are these guys smoking?

• This is BS - wait for the Poll Troll to deconstruct their farkakte theory.

  DL 2012 Poll Troll troll

• Much of this ground has already been covered by fivethirtyeight.com. Nate Silver has noted that traditional economic models, which draw heavily on past experience rather than on observation of campaigns, just don't seem to be working this year. Still, maybe that too is an illusion; we shall see.

• This model was largely discounted by Nate Silver last night, OP. [quote]A Denver Post reporter asked me about this U. of Colorado election model [quote]It's late, so I'll be blunt: I saw their paper and I think there are glaring problems with their methodology. [quote]The U. of Colo. model fits the equivalent of 7 unknowns to 8 elections. That's not a good idea. [quote]The Colo. model also assumes huge effects from unemployment if incumbent is a Dem., but none if he's GOP. Hard claim to defend. [quote]If you want a "fundamentals" model that shows Romney winning, the Hibbs model is a lot more sensible. [quote]Also, it's false advertising to claim CU model has predicted the last 8 elections right. It's a new model. Hasn't predicted anything yet.

  https://twitter.com/fivethirtyeight

  2012%20Poll%20Troll

• thank you poll troll!

• If this happens it will be a fucking landslide for the GOP! Either these guys, or Nate Silver are wrong. Hope someone remembers to bump this thread on Election night.

• Utter shit. I am a researcher working in sampling and statistics and surveys. Their methodology is deeply flawed. As a matter of fact, the robocalls and the landline bases of all (to my knowledge) the polls showing good Romney results do not take into account the higher proportion of Democrats using non-traditional communication technology. There is an immediate undercount of Obama votes in all of these results, and it is most serious in the swing states where diverse populations and more-affluent, voting, informed citizens reside who are more likely NOT to vote for Romney under any circumstances. Ignore the every-other-week claims by "academics" you never heard of. It's propaganda.

  Researcher%20in%20the%20Field

• i heart r7. typed on my iPhone!

• R7, You so eloquently point out what I've been trying to communicate. Do you think that this new poll illustrates "scare tactics" by the pro-Romney camp, because they are down in almost every traditional poll, they didn't get as good a bounce as they had wanted from their VP choice despite massive spending on ads, and their crucial candidate likability ratings are far worse than Obama's? And of course every statement from their camp is totally ridiculous? Waiting to be dissed by the trolls. . .

• Limbaugh is going batshit over this today. He's declaring these guys to be the geniuses and saying this confirms what his gut has been telling him all along (insert your own Limbaugh gut joke here).

• People will go anything for money. Especially statisticians.

• Who paid for their study? That is the question you must always ask.

• Anything we can get out there to make the GOP complacent and fire up the Dems... But the article is ridiculous...read the caveats. They're working with incomplete data sets... Duh.

• ANOTHER FREEPER THREAD WITH A BANNER THAT THEY WANT TO TURN INTO A MEME. NO REASON TO GO ON HERE.

• I wonder what they would say now to Gawker's new revelations? Better get another "expert" to predict MR success, before the campaign $$ donations dry up for the Repubs.

• "ANOTHER FREEPER THREAD..." What the freeper fails to realize is that these Mitt Will Win poll numbers actually rally us. We will give up tickets to Cher's 16th farewell tour, buy cheaper vodka, skip a month at the dog groomer and lube with only spit, in order give money to blunt Mitt's winning.

• Darling, we do not scream. Ever.

• "oh well....I guess I just better stay home and not VOTE AT ALL..."

  *wink wink*

• No we don't, R17. Except once, when Princess Diana died and we were all sitting there watching when it was announced. R9, the academics involved appear to be sincere and qualified, as far as that goes. But they shouldn't presume where better researchers have methodology more firmly in hand. By the way, even the best pollsters (and I'm in agreement) declare that this is not a cakewalk for Obama, that voters are upset and unhappy, and that unexpected events could overtake either side. Look at Akin in Missouri (where I live). A mere Midwest Senate race has provided unexpected, complex dynamics (not all against the Republicans) that confuse and alter the trend lines. With that said, it is clear that Obama has the advantage. ANY poll that shows a Romney uptick will be shouted from the rooftops by Republicans because they seek to break any sense of Obama's inevitability, and make it look like the swing states are more fluid than they likely are. Democrats NEVER are as craven as the Republicans are in critical way. They fight and fight dirty, but they - still sensing that they represent the majority of the country, albeit from a leftward position - have limits to their reprehensibility. Usually. Republicans, with the still-ingrained sense of minority zealotry, the true-believer stance of the only real America that Nixon first perfected and Reagan brought to its apotheosis, will do ANYTHING to win. And now they have a presumptive nominee who will take any position on any topic necessary, because he represents a minority within a minority - a true believer within the true believers. The fact that the party has devolved to the Luddites of the Tea Party - the sop of Ryan as a last-gasp try for support by Romney, who loathes the Tea Baggers - show his lack of dignity. That's the context for all this. My meta-take on all trends is that the election is Obama's to lose. No economic news is likely to change things for the worse for him, state by state. And once Romney opens his mouth and Ryan's positions are better known, the trickle back to the Democrat side should keep almost all swing states in the Democrat's column. Unless something unexpected happens. I think Florida is the one real toss-up. I see everything else, including Wisconsin, going with Obama, and the Democrats ending with at least 203 electoral votes. But ignore the second stringers who declare they have gotten everything right since 1980. Survey methodology is completely different now than then, and even the best of us have not yet come up with a sure-fire model for overcoming the difficulties. Nate Silver deserves everyone's attention, if not daily, almost.

  R7

• I live in Minnesota. There is no way in hell Romney is winning here.

• [quote]As a matter of fact, the robocalls and the landline bases of all (to my knowledge) the polls showing good Romney results do not take into account the higher proportion of Democrats using non-traditional communication technology. THIS. Who still has a landline as their primary phone, and also who answers numbers they do now know on their caller ID? OLDER PEOPLE! And of course older people are more likely to vote Republican. I know very few people under 30 or even 35 who even have a landline. They only have a cell, and many of them will not answer a number they do not know, and let it go to VM. Poll results in this day and age are not as reliable as they were years ago.

• [quote]Except once, when Princess Diana died and we were all sitting there watching when it was announced. Outliers.

• I'm seeing a bit of a trend in polling that shows the GOP is more "excited" about the election. That can't be good news for election day.

• Haa haaha haaa!! Link, r23, or you're just a crappy troll.

  %28as%20if%20we%20don%27t%20know.%29

• What they're not telling you is that their model takes into account voter suppresion, voter ID, voter roll purging, and rigging the vote counting. Be ever vigilent.

• [quote]Link, [R23], or you're just a crappy troll. Generally speaking, I think R23 is right, that there has been a bit of an "enthusiasm gap" in this election. Not nearly as bad as 2010, though. Below is a balanced look at this from a left-leaning news site.

  http://2012.talkingpointsmemo.com/2012/08/republican-voter-enthusiasm.php

• It's worth noting that they aren't really all that enthusiastic about voting *for* Romney; they're enthusiastic about voting *against* Obama.

• Florida isn't a tossup now with Ryan's Medicare issue and Rommey's appeal to only 28% of Latino voters.

• I'm in Ohio where all 88 counties have pretty much been secured for Romney by the Secretary of State. I'm honestly contemplating suicide if Romney wins. Maybe I'll wait a month to see if the Aztec prophecy comes to pass, but I can't stand any more waiting for America to show some common sense. I'm already depressed from a death in the family and frankly it's past my time.

• [quote]Florida isn't a tossup now with Ryan's Medicare issue That's not necessarily true. While it's true that the overwhelming majority of voters hate the Ryan plan once it's explained to them, Romney and Ryan are blatantly lying about both their plan and the President's plan, the so-called "716 billion dollar cut to Medicare to pay for Obamacare." At the moment, those lies are working with a pretty sizable percentage of the electorate. Obama needs to find a way to cut through the bullshit on that and on the welfare-to-work issue that is also a blatant lie. If he can do so, I think he wins. If he cannot, it's a tossup and Romney may well win.

• Scienticians have way more credibility than these two.

• People from Boulder, CO never lie.

  Mrs.%20Patsy%20Ramsey%2C%20formerly%20of%20Boulder%2C%20CO

• r25 Exactly now that Obama is letting illegals vote, he's sure to win the election. r27 And that is how Obama got elected. People were voting against the Republicans

• [quote]Exactly now that Obama is letting illegals vote, he's sure to win the election. At what point do the lies and hyperbole stop, you worthless trash? There are a lot of idiots in this country who shouldn't even be allowed to vote because they're so God damned stupid... and they elect morons like Todd Akin.

• Will someone ask their grandma to shit in my mouth? PLEASE?

  R33

• [quote]Exactly now that Obama is letting illegals vote, he's sure to win the election. Wow. You are one dumbass white trash closet case.

• [quote]Exactly now that Obama is letting illegals vote, he's sure to win the election. The stupid, it burns.

• I seem to remember in 2000 virtually every model said Gore would beat Bush, it did not quite turn out that way. I would add the Colorado Uni survey was probably no different, it says it predicted in 2000 accurately when Gore won the popular vote and Bush the electoral college, ie it probably predicted a Gore win too!

  HYUFD

• 1. If you read Nate Silver's comments from the past week, he has said that this Colorado electoral model has multiple problems with its methodology. 2. A lot has changed demographically since 2000. The Electoral Map has shrunk for the Republicans and they are not able to win as many states as they used to. For example, George W. Bush was able to win New Mexico in one of his elections. Fast forward to 2008 & 2012 and New Mexico is in Obama's column. Nevada is also trending away from the Republicans, as is New Hampshire. Possibly Ohio too. And Virginia is now a state that the Democrats can win (as is North Carolina). Another change from 2000/2004.

  2012%20Poll%20Troll

• R38 - This model didn't actually predict anything accurately before anyone voted. It's a new model that claims to be predicting past election outcomes accurately. How hard is that really?

• R33, pathetic assertion 2: and the fact that McCain is a bumbling old fool with a predilection towards pretty girls, and those girls being as dumb as as a sack of rocks, had nothing to do with it.

• God I wish we would ditch this fucking outdated Electoral College once and for all and go to a full popular vote. This is bullshit that Ohio picks a President for the whole fucking country.

• I think the Poll Troll has accurately dissected the fallacy in this model, but I'll also note that Nate Silver still has Obama's chances of winning at over 70%, and HIS model actually HAS been proven to work with the changing dynamics of polling. He was almost the ONLY one who accurately predicted the scope of Obama's 2008 win, and correctly predicted him winning a number of states that every other pollster thought he had no chance of winning, including Virginia and North Carolina.

• Here ya go, R6!

• So, how'd that work out for you?

• OOPS

• Those pieces of shit deserve to be denied tenure.

• BWAHHAHHAHHA!

• BOOM!